Diabetic prediction framework using optimisation strategy via optimal weighted score-based deep ensemble network to support diabetic patients
by Santosh Kumar Bejugam; Jyothi Vankara
International Journal of Bioinformatics Research and Applications (IJBRA), Vol. 19, No. 5/6, 2023

Abstract: Diabetes is one of the dangerous diseases that increase blood glucose levels, and it affects the patient's life. Next, in the deep feature extraction stage, the collected data is employed as the input. Here, the deep features are extracted using one-dimensional convolutional neural network (1DCNN). Then, the acquired optimal features are offered as the input to intelligent deep ensemble network (IDENet) that holds the networks such as long short-term memory (LSTM), 1DCNN, deep temporal context networks (DTCN) and extreme learning (EL). The parameters of IDENet are tuned by enhanced light spectrum with horse herd optimisation (ELS-HHO). Further, the attained predicted values from the IDENet are fed as the input to the weighted fusion of predicted values. Then, their weights are tuned by ELS-HHO to attain the effective glucose prediction outcome. Finally, the suggested glucose prediction model secured a better prediction rate than the classical glucose prediction models in experimental observation.

Online publication date: Fri, 14-Jun-2024

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Bioinformatics Research and Applications (IJBRA):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com